February 2004

Thursday 5th February 2004

I decided to investigate the twitchiness of the robot following the Worthing event. The radio reception of Hog 2 had never been quite as good as Hog 1, but I hadn't as yet figure out why this was. I had recently been on the Technobots website to find that there was a new addition to the technical guides section that talked about interference problems and how to resolve them. Encourage by the very interesting and useful guide notes from Tim Mann, I started to look seriously into my own problem.

The first thing I checked was whether the twitchiness was due to all the power supplies to the controller being generated from my single 24 volt battery supply. I connected an independent set of batteries to the receiver and then turned on the transmitter with the aerial retracted. I didn't have to walk far away from the Hog before the motors twitched. I didn't know if the interference was there all the time or only when a valid radio signal was detected, so I connected a servo to a spare port on the receiver and watched it as I turned the transmitter on and off. I noticed that with the receiver off, and consequently the failsafe relays on the robot turned off, the servo was steady. As soon as I turned the transmitter on and the failsafe relay switched power through to the motors and actuators, it was then that the servo twitched as well as the motors.

I prodded about with the oscilloscope probe for quite a while trying to spot anything obvious, but couldn't see anything. I soldered a few capacitors in strategic places to smooth out power rails, but to no avail. I then, by chance, looked at the input feed to the DC-DC converter that I use to generate a 36v supply to drive the high-side MOSFETs. To my amazement there were horrible 1 volt spikes continuously on the input. This 12 volt input feed to this converter came from a power regulator that I had forgot to smooth out properly with a capacitor, so I quickly soldered a 100uF capacitor across it and checked it again. To my relief it made a world of difference, so I was well chuffed with the stability of the controller by the end of the evening.


Go to top top
 

Thursday 15th February 2004

On 14th and 15th February I took part in the the Brighton Modelworld exhbition and had a great time. You can read about what happened in the events section.

With the exhbition now over, I had learnt a lot because of the number of battle I had endured. As a result I realised there were a number of improvements that I could make; some as a result of comments made during the tech-check' and others from the battle injuries sustained throughout the event.

The things I think need to concentrate on immediately are:

Gas bottle protection

There needs to be some form of protection for the gas bottle once the flipper is up. This only needs to be a layer of polypropylene covering the exposed front facing part of the battle.


Gas bottle access flap

There is still some ambiguity over access to the gas bottle tap so that you can turn it off in an emergency without the need of a tool. I will be looking into cutting a whole in a panel and fitting a flap over the top.


More panel fixing bolts

To keep the weight down to a minimum, I have used the minimum number of bolts possible to secure the panels to the chassis. I think I need to fit more bolts so that a well placed blow from another robot doesn't completely dislodge a panel, but having more fixing points will also make the structure a lot more rigid and sturdy. I will also use longer and thicker bits of metal as fixing tabs so they can take more knocks.


Allen key panel fixing bolts

I currently use slot head screws to fix the panels to the chassis, but the slots tend to get damaged too easily, which makes it difficult to undo them with a screwdriver after that. I think I will now use Allen keys even though they are quite a bit more expensive.


24volt motors, not over-run

I currently use 12v 630Watt Bosch motors that I overrun at 24volts on the principle that I can get four times the power out of them. What I am finding is that I don't have the friction with the tyres to take advantage of this extra power, and that the motors draw four times the current than they would normally. All this means that I stress the motor controller a lot when the motors stall or are under high load, resulting in the controller blowing a few MOSFETs every now and then. The expensive option is to replace these 12 volt motors with the more commonly used 24 volt 750 Watt Bosch motors. I don't know if I can justify this expense at the moment, so this improvement may take a back seat for a while.


Smaller drive sprockets on wheel axles

I use fairly large sprockets on the drive shafts to get a reasonable gear ratio between the motor and wheels, but this means that the sprockets protrude below the chassis of the robot. I am therefore susceptible to the sprockets getting damaged if I ride over the top of other robots or obsticles, or I get flipped over and rolled about. It may be worth the trade off to slightly reduce the gear ratio so that the sprockets not longer protrude below the chassis.


Battery capacity

I was finding that my batteries were not really lasting the full battle period, which tended to leave me immobilised at the end of a long battle. Part of the reason was that the battles were coming so frequently that my chargers didn't have the time to fully recharge the batteries before the next battle was due. Larger batteries might help me, but smaller motors too might be a good idea too. Over running the 12 volt motors at 24 volts means I am using four times the power. Using larger batteries has a weight penalty, so the best but more expensive solution might be to change the motors!

I've signed up to go the the Portsmouth exhibition in May, so will have to start on some of these improvements over the coming weeks.

Go to top top
Go to top top

Back to Hassocks Hog home pageBack to January 2003Forward to May 2004

Last updated 9th May